run-train
Rigor Train skill for deep learning research repositories. Use when a documented or selected training command should be run conservatively for startup verification, short-run verification, full kickoff, or resume, with command, config, seed, log, checkpoint, status, and metric evidence written to standardized train_outputs/. Do not use for environment setup, exploratory sweeps, speculative idea implementation, or…
- installs 8w
- 159,991
- 30-day movement
- starts with the next reading
- Related entries
- 1
- Connections
- 0
run-train is a Claude Code skill that conservatively executes an already-selected training command in a deep learning research repository, in modes like startup verification, short-run verification, full kickoff, or resume. It writes standardized evidence—command, config, seed, logs, checkpoints, status, and metrics—to a `train_outputs/` directory including SUMMARY.md, COMMANDS.md, LOG.md, SCIENTIFIC_CHANGELOG.md, and COMPARABILITY_REPORT files.
It gives you disciplined, evidence-logged training runs with clear boundaries instead of ad-hoc command execution.
Use it to
- Verify a training script starts correctly before a full run
- Run a short smoke training run to validate the pipeline
- Kick off a full documented training run with standardized evidence
- Resume an interrupted training run with consistent logging
For Researchers running deep learning training in reproducible repositories
- Host repository
- lllllllama/RigorPilot-Skills
- Installs, lifetime
- 311k
- Installs, 8 weeks
- 160k
- Host stars
- 489
- Host language
- Python